Geology Reference
In-Depth Information
compensate for minor amounts of contaminants in the silicon. These detectors must be cooled with liquid nitrogen
to enhance their performance and prevent the migration of the lithium out of the crystal sensor. When an X-ray
photon strikes the detector, it generates an output signal that is proportional to the energy of the X-ray photon. A
multichannel analyzer sorts output signals into different energy ranges and plots a histogram of the X-ray energies
detected over time (Figure 10.1.4). Most EDS detectors have a Be (beryllium) window that is strong enough to
withstand the pressure change when the vacuum column is vented to the atmosphere. The window protects the
detector by maintaining the chamber vacuum, but it also absorbs X-rays from elements lighter than Na so that light
elements cannot be detected. Some EDS detectors are equipped with thinner windows or are windowless by
continuous chamber evacuation strategies. These detectors can detect X-rays emitted by light elements. EDS
spectra provide a rapid way to qualitatively identify the elements present in a sample. An X-ray spectrum, i.e., an
energy distribution graph, can be acquired in just a few seconds and with a little experience, the analyst can quickly
identify the different phases present in a sample. Newer electron probes are equipped with silicon drift detectors
(SDDs), which are photodiodes. These are similar to EDS detectors but do not require liquid nitrogen for cooling,
but use reverse electronic bias to thermally cool the sensor, thus they can process at much higher count rates, which
makes them capable of detecting light elements.
WDS detectors contain an X-ray proportional counter and a diffracting crystal of known d-spacing, i.e., the spacing
between atomic planes. The counter and crystal are moved along the circumference of a goniometer-focusing circle
to satisfy Bragg
s law for the X-ray of interest. Unlike an EDS detector, a WDS detector is only tuned to one
element at a time. Most microprobe spectrometers contain two or four diffracting crystals and are configured so that
most of the X-ray spectrum can be detected.
'
Each X-ray detector chosen for a project has strengths and weaknesses that must be considered in order to determine
which detector best satisfies the requirements of the project. EDS detectors are fast, relatively inexpensive, and easy to
maintain, and can detect and measure X-rays from rough surfaces, but they have several disadvantages. Since the
EDS detector looks at a wide range of the X-ray spectrum simultaneously, the background signal is fairly high, and it
cannot detect concentrations below about 0.5wt% (5000 ppm). EDS detectors also have relatively poorer energy
resolution than a dedicated wavelength energy detector. Many elements have such similar characteristic X-ray
energies that the EDS detector cannot distinguish between them. For example, the emission energy of the S K
α
peak is 2.308 keV and of the Pb M
peak is 2.346 keV. A peak in this energy range could indicate the presence of
elemental sulfur, elemental lead, or PbS, which is the mineral galena. This is a particularly important concern if peaks
are misidentified by automated vendor-supplied software. The EDS detectors also generate several artifacts that must
be considered when interpreting EDS spectra. Escape peaks occur when the SiLi detector absorbs an X-ray photon
and emits a Si photon. The output X-ray will generate a small peak with energy of 1.74 keV lower than the original
X-ray photon. This can also result in the presence of a small spurious silicon peak when, in fact, no silicon is present.
Overlapping peaks result when two different X-ray photons arrive at the detector so close together that the detector
cannot distinguish them. The result is a peak with energy that is the sum of the two contributing photons and can
indicate the presence of elements that are not present in the sample.
α
WDS Spectrometers are large and mechanically complicated. However, they have much better energy resolution
than EDS detectors and can usually distinguish most peak overlaps that plague EDS detectors. WDS detectors only
sample a narrow portion of the X-ray spectrum at a time, so they have a much lower background signal and in some
cases can detect elemental concentrations as low as 100 ppm. Since WDS detectors only look at one element at a
time, the analyses of multiple elements must be done sequentially and therefore take longer. An example of
mapping a particular element of interest with a WDS detector is shown in Figure 10.1.5. The center figure shows
tiny salammoniac crystals covered with sulfur. The adjacent WDS maps (X-ray maps) to the left and right of this
figure show the location of Cl and S in the minerals. WDS detectors are also much less forgiving of fragile mineral
samples like those found at coal-fire gas vents. Optimum quantitative analyses require samples that are flat and
polished. Fragile materials like many coal-fire residues may not survive the rigors necessary to produce a polished
section for optimum analytical results.
The capabilities of XRD and the electron microprobe are complimentary. Although an XRD pattern is
sensitive to the kinds of atoms in a mineral, some minerals are isostructural. Such minerals have the same
crystal structure, but they contain some different kinds of atoms. Consequently, their diffractogram patterns
may be very similar. EMPA provides a means to quickly determine what elements are present in a sample and
to quantitatively measure the sample
s composition. Backscattered and secondary electron imaging document
compositional and morphologic variations in a sample, and WDS mapping documents the distribution of
elements in a sample.
'
